Search Strategy

A search strategy is a structured, documented plan for identifying relevant studies in bibliographic databases and other sources, specifying the databases to be searched, the search terms (including subject headings and free text words), the Boolean operators connecting terms, and any filters or limits applied. In systematic reviews, the search strategy must be comprehensive, transparent, and reproducible.

The quality of a systematic review depends critically on the comprehensiveness of its search. A search that misses relevant studies produces a biased and incomplete evidence synthesis. Conversely, a search that is too broad retrieves an unmanageable number of irrelevant records, wasting screening time. A well constructed search strategy balances sensitivity (capturing all relevant studies) and precision (minimizing irrelevant results). Reporting the full search strategy allows readers and peer reviewers to evaluate its adequacy and enables replication. A strategy that is not reported in full cannot be reproduced or critiqued by readers.

Developing a search strategy begins with breaking the research question into its component concepts, typically using the PICO framework. For each concept, the researcher identifies relevant subject headings (such as MeSH terms in PubMed or Emtree terms in Embase) and free text synonyms, including variations in spelling, abbreviation, and terminology. Terms within each concept are combined using the OR operator, and the concepts are combined using the AND operator.

The strategy is then adapted for each database, as subject heading vocabularies and search syntax differ across platforms. AMSTAR 2 expects at least two databases relevant to the question, and for Cochrane intervention reviews the Cochrane Handbook asks for CENTRAL, MEDLINE and Embase (if the team has access) as a minimum, plus trials registers and subject specific databases where relevant. Supplementary search methods include scanning reference lists of included studies, forward citation searching, hand searching key journals, searching clinical trial registries, contacting experts in the field, and searching for grey literature (conference abstracts, dissertations, and government reports).

The PRESS (Peer Review of Electronic Search Strategies) checklist provides a structured framework for peer reviewing search strategies, focusing on translation of the question into search concepts, term selection, Boolean logic, syntax, and filter use. The Cochrane Handbook recommends that search strategies be peer reviewed by a librarian or information specialist before they are run.

Search filters can be applied to limit results by study design (such as the Cochrane Highly Sensitive Search Strategy for RCTs), language, date range, or geographic setting. However, filters can inadvertently exclude relevant studies and should be used cautiously with their limitations acknowledged.

Develop a formal search strategy for every systematic review, meta analysis, and scoping review. A draft of the strategy, for at least one database, should be documented in the review protocol before searches are conducted, and later changes reported. Report the full search strategies for all databases, registers and websites searched, including any filters and limits, in the manuscript or supplementary materials. For narrative reviews and primary research literature reviews, a less formal search may be acceptable, but documentation of sources and terms searched improves transparency.

A common error is searching only one or two databases, missing studies indexed elsewhere. Another mistake is relying solely on text word searching without using database specific subject headings, which can miss studies that use different terminology. Researchers sometimes apply overly restrictive filters (such as English language only) without acknowledging the limitations. Failing to search grey literature and trial registries is a significant omission that increases the risk of publication bias. Finally, not documenting the search strategy in sufficient detail to allow replication undermines the transparency and reproducibility of the review.

A systematic review on the effectiveness of group cognitive behavioral therapy for social anxiety develops search strategies for PubMed, Embase, PsycINFO, CINAHL, and the Cochrane Central Register of Controlled Trials. The PubMed strategy includes MeSH terms ("Cognitive Behavioral Therapy," "Phobia, Social") combined with free text synonyms ("CBT," "cognitive behav*," "social anxiety," "social phobia") using Boolean operators. The search is limited to randomized controlled trials using the Cochrane Highly Sensitive Search Strategy filter. The strategy is peer reviewed using the PRESS checklist by a health sciences librarian, who identifies three missing synonyms that are added before the final search.